10 citations · 23 across the 26 of their papers we have counts for
30 papers
PCA-Seg: Revisiting Cost Aggregation for Open-Vocabulary Semantic and Part Segmentation
Jianjian Yin, Tao Chen, Yi Chen +4
Recent advances in vision-language models (VLMs) have garnered substantial attention in open-vocabulary semantic and part segmentation (OSPS). However, existing methods extract ima…
Beyond Gemini-3-Pro: Revisiting LLM Routing and Aggregation at Scale
Shengji Tang, Weihao Lin, Peng Ye +9
Large Language Models (LLMs) have rapidly advanced, with Gemini-3-Pro setting a new performance milestone. In this work, we explore collective intelligence as an alternative to mon…
Wisdom of the Crowd: Reinforcement Learning from Coevolutionary Collective Feedback
Wenzhen Yuan, Shengji Tang, Weihao Lin +8
Reinforcement learning (RL) has significantly enhanced the reasoning capabilities of large language models (LLMs), but its reliance on expensive human-labeled data or complex rewar…
A Scalable Multi-LLM Collaboration System with Retrieval-based Selection and Exploration-Exploitation-Driven Enhancement
Shengji Tang, Jianjian Cao, Weihao Lin +7
Existing multi-LLM collaboration systems often encounter scalability challenges when integrating new LLMs and tasks, leading to suboptimal performance. To address this, we propose…
PaceLLM: Brain-Inspired Large Language Models for Long-Context Understanding
Kangcong Li, Peng Ye, Chongjun Tu +6
While Large Language Models (LLMs) demonstrate strong performance across domains, their long-context capabilities are limited by transient neural activations causing information de…
Decouple and Orthogonalize: A Data-Free Framework for LoRA Merging
Shenghe Zheng, Hongzhi Wang, Chenyu Huang +5
With more open-source models available for diverse tasks, model merging has gained attention by combining models into one, reducing training, storage, and inference costs. Current…
Dynamic Base model Shift for Delta Compression
Chenyu Huang, Peng Ye, Shenghe Zheng +4
Transformer-based models with the pretrain-finetune paradigm bring about significant progress, along with the heavy storage and deployment costs of finetuned models on multiple tas…
Breaking the Compression Ceiling: Data-Free Pipeline for Ultra-Efficient Delta Compression
Xiaohui Wang, Peng Ye, Chenyu Huang +5
With the rise of the fine-tuned-pretrained paradigm, storing numerous fine-tuned models for multi-tasking creates significant storage overhead. Delta compression alleviates this by…
FAVOR-Bench: A Comprehensive Benchmark for Fine-Grained Video Motion Understanding
Chongjun Tu, Lin Zhang, Pengtao Chen +5
Multimodal Large Language Models (MLLMs) have shown remarkable capabilities in video content understanding but still struggle with fine-grained motion comprehension. To comprehensi…
TokenCarve: Information-Preserving Visual Token Compression in Multimodal Large Language Models
Xudong Tan, Peng Ye, Chongjun Tu +5
Multimodal Large Language Models (MLLMs) are becoming increasingly popular, while the high computational cost associated with multimodal data input, particularly from visual tokens…
Attention Reallocation: Towards Zero-cost and Controllable Hallucination Mitigation of MLLMs
Chongjun Tu, Peng Ye, Dongzhan Zhou +4
Multi-Modal Large Language Models (MLLMs) stand out in various tasks but still struggle with hallucinations. While recent training-free mitigation methods mostly introduce addition…
Seeing Delta Parameters as JPEG Images: Data-Free Delta Compression with Discrete Cosine Transform
Chenyu Huang, Peng Ye, Xiaohui Wang +5
With transformer-based models and the pretrain-finetune paradigm becoming mainstream, the high storage and deployment costs of individual finetuned models on multiple tasks pose cr…
DeRS: Towards Extremely Efficient Upcycled Mixture-of-Experts Models
Yongqi Huang, Peng Ye, Chenyu Huang +5
Upcycled Mixture-of-Experts (MoE) models have shown great potential in various tasks by converting the original Feed-Forward Network (FFN) layers in pre-trained dense models into M…
Multi-Level Decoupled Relational Distillation for Heterogeneous Architectures
Yaoxin Yang, Peng Ye, Weihao Lin +4
Heterogeneous distillation is an effective way to transfer knowledge from cross-architecture teacher models to student models. However, existing heterogeneous distillation methods…
All-in-One: Transferring Vision Foundation Models into Stereo Matching
Jingyi Zhou, Haoyu Zhang, Jiakang Yuan +5
As a fundamental vision task, stereo matching has made remarkable progress. While recent iterative optimization-based methods have achieved promising performance, their feature ext…
S2HPruner: Soft-to-Hard Distillation Bridges the Discretization Gap in Pruning
Weihao Lin, Shengji Tang, Chong Yu +2
Recently, differentiable mask pruning methods optimize the continuous relaxation architecture (soft network) as the proxy of the pruned discrete network (hard network) for superior…
HiSplat: Hierarchical 3D Gaussian Splatting for Generalizable Sparse-View Reconstruction
Shengji Tang, Weicai Ye, Peng Ye +4
Reconstructing 3D scenes from multiple viewpoints is a fundamental task in stereo vision. Recently, advances in generalizable 3D Gaussian Splatting have enabled high-quality novel…
DynaSurfGS: Dynamic Surface Reconstruction with Planar-based Gaussian Splatting
Weiwei Cai, Weicai Ye, Peng Ye +2
Dynamic scene reconstruction has garnered significant attention in recent years due to its capabilities in high-quality and real-time rendering. Among various methodologies, constr…
FNP: Fourier Neural Processes for Arbitrary-Resolution Data Assimilation
Kun Chen, Tao Chen, Peng Ye +5
Data assimilation is a vital component in modern global medium-range weather forecasting systems to obtain the best estimation of the atmospheric state by combining the short-term…
-DiT: A Training-Free Acceleration Method Tailored for Diffusion Transformers
Pengtao Chen, Mingzhu Shen, Peng Ye +5
Diffusion models are widely recognized for generating high-quality and diverse images, but their poor real-time performance has led to numerous acceleration works, primarily focusi…
EMR-Merging: Tuning-Free High-Performance Model Merging
Chenyu Huang, Peng Ye, Tao Chen +3
The success of pretrain-finetune paradigm brings about the release of numerous model weights. In this case, merging models finetuned on different tasks to enable a single model wit…
Once for Both: Single Stage of Importance and Sparsity Search for Vision Transformer Compression
Hancheng Ye, Chong Yu, Peng Ye +5
Recent Vision Transformer Compression (VTC) works mainly follow a two-stage scheme, where the importance score of each model unit is first evaluated or preset in each submodule, fo…
Enhanced Sparsification via Stimulative Training
Shengji Tang, Weihao Lin, Hancheng Ye +4
Sparsification-based pruning has been an important category in model compression. Existing methods commonly set sparsity-inducing penalty terms to suppress the importance of droppe…
MADTP: Multimodal Alignment-Guided Dynamic Token Pruning for Accelerating Vision-Language Transformer
Jianjian Cao, Peng Ye, Shengze Li +4
Vision-Language Transformers (VLTs) have shown great success recently, but are meanwhile accompanied by heavy computation costs, where a major reason can be attributed to the large…